Videoconferencing services may enable two or more locations to simultaneously interact via two-way audio and video transmissions. In a typical videoconferencing session, two or more end-devices, that may each include a video display device and a video camera, may connect, such as through an Internet Protocol (IP) based communication session, either to one another or to a bridge device. Video from each end-device may be compressed, transmitted as an IP-stream over a network, and provided at the other end-device(s).
A videoconference, and particularly a high definition videoconference (HDVC), may be a relatively bandwidth intensive service. Accordingly, videoconferencing services may require high investment in network infrastructure by the service provider. Ideally, a network that supports videoconferencing may include a combination of features such as high capacity, low latency/delay in the network elements to allow for interactive conversations, and little or no packet loss or network errors.
A number of attempts to address the difficulties associated with providing HDVC services on a network have been tried. One solution may be to build a high-capacity (over provisioned) network to accommodate 100% of peak demand for the HDVC services at all times. This approach can be generally expensive and may lead to low utilization of network infrastructure during off-peak hours. Another solution may be to implement static Quality of Service (QoS) on the network. This approach can be inadequate as it may be difficult to deliver consistent service quality during busy periods and may generally result in an underutilized network during off-peak hours.